What token usage should I aim for when building an AI agent? I see a mismatch between 3K tokens in logs and 71.2K in the editor.
I don’t sell AI agents, but running one is cheap: unless you’re using the most expensive model (GPT‑4.5) and paying for everything yourself, the cost is marginal. With cheaper models like GPT‑4.1 mini, you won’t notice the expense unless you’re being spammed constantly.
Honestly nowadays I'm paying for most of the tokens just because it's so cheap. It makes zero difference to me and it just minimizes friction. If you want to be complete, you'd also get them to sign up for an OpenAI account. Keep in mind there are nuances like rate limits; some clients don't have high rate limits but want a high‑rate‑limit application built quickly. I usually use my own OpenAI API key.
They’re essentially free to run—about $20 a month plus any token usage. If you want to replicate the setup, just follow one of my courses from start to finish. The agent is just the harness with connectors and context for your environment. Platforms like Grockbot abstract away credential management, making it easy to connect everything and unlock more value.
Nate explains that pricing should be proportional to the value the system provides. He starts by considering where you are in your journey and what you’re optimizing for—brand, case studies, testimonials can let you charge more. If you’re just starting, focus on gaining experience and understanding the exact value you deliver, which you need to express in terms the client understands: hours saved and dollars saved. In discovery, ask the right questions to get those figures. Then make it a no‑brainer: for example, a system saving 5 hours a week when the client values an employee’s time at $100/hour saves $500 weekly; over months that could be $50k, so you could charge $7–$15k (or even more) because the client is saving $35k. You have to frame the benefit in the client’s language at a high level. The other speaker agrees, saying Nate hit on many important concepts—packaging of the service, the provider, perceived and objective value. They then shift to a quick sponsor shout‑out for Whimo, noting how convenient and cost‑effective it is (about 30% more than an Uber but 10× safer), mention that their Instagram ads have been only for School and Whimo, and wrap up by thanking each other, saying it was a pleasure, looking forward to hanging out again, and promising to cover brand stats tomorrow.
Higgs field cloud combined with MCP and unlimited tokens is currently an alpha advantage. It lets you do high‑quality creative work for companies that still run old‑school, manual ad flows that are bottlenecked or throttled. Because those companies can’t test quickly, stepping in with this kind of automation gives you a big edge. I’ve been making videos about this approach for about a month, so it’s definitely worth experimenting with.